Brittle and Ductile Behavior in Deep-Seated Landslides: Learning from the Vajont Experience

Abstract

This paper analyzes the mechanical behavior of the unstable Mt. Toc slope before the 1963 catastrophic collapse, considering both the measured data (surface displacements and microseismicity) and the updated geological model of the prehistoric rockslide. From February 1960 up to 9 October 1963, the unstable mass behaved as a brittle-ductile 'mechanical system,' characterized by remarkable microseismicity as well as by considerable surface displacements (up to 4-5 m). Recorded microshocks were the result of progressive rock fracturing of distinct resisting stiff parts made up of intact rock (indentations, undulations, and rock bridges). The main resisting stiff part was a large rock indentation located at the NE extremity of the unstable mass that acted as a mechanical constraint during the whole 1960-1963 period, inducing a progressive rototranslation toward the NE. This large constraint failed in autumn 1960, when an overall slope failure took place, as emphasized by the occurrence of the large perimetrical crack in the upper slope. In this circumstance, the collapse was inhibited by a reblocking phenomenon of the unstable mass that had been previously destabilized by the first reservoir filling. Progressive failure of localized intact rock parts progressively propagated westwards as a consequence of the two further fillingdrawdown cycles of the reservoir (1962 and 1963). The characteristic brittle-ductile behavior of the Vajont landslide was made possible by the presence of a very thick (40-50 m) and highly deformable shear zone underlying the upper rigid rock mass (100-120 m thick).

The Vaiont landslide: re-assessment of the evidence leads to rejection of the consensus

“…2) within a broad bowl- The failure was a reactivation of an ancient landslide. Guidici and Semenza (1960) 1 , Semenza (1965), Voight and Faust (1982), Hendron and Patton (1985), Belloni and Stefani (1992), Nonveiller (1992), Tika and Hutchinson (1999), Mantovani and Vita-Finzi (2003), Ward and Day (2011), Superchi (2012), Hungr and Aaron (2013), Bolla (2012, 2015), Paronuzzi et al (2016), Crosta et al (2016), Zhao et al (2016), Pasuto (2017) C2 Both the prehistoric landslide and the 1963 landslide took place along thin clay seams in a limestone mass and that these were at, or near, residual strength prior to 1960. Semenza (1965), Skempton (1966) 2 , Müller (1968) 3 , Voight and Faust (1982), Hendron and Patton (1985), Tika and Hutchinson (1999), Ward and Day (2011), Hungr and Aaron (2013), Zhao et al (2016)…”

“…The trigger for the failure was a major loss of stability due to inundation of the toe of the slide mass by the impounded reservoir. Kiersch (1964), Semenza (1965), Chowdhury (1978), Hendron and Patton (1985) 8 , Ward and Day (2011), Delle Rose (2012), Bolla (2012, 2015), Paronuzzi et al (2016) C1-C4 All of the above. Hutchinson and Kwan (unpubl.1986), Hutchinson (1987) 9 , Alonso et al (2010), , Pinyol and Alonso (2010), Del Ventisette et al (2015) 10 , Havaej et al (2015) 11,12 , Wolter et al (2016 1 They suggested the presence of an ancient landslide prior to the failure event 2 Due to flexural slip rather than a previous landslide 3 Not as 'residual strength' but as 'unusually low friction values' (ϕ p ) 4 Less extreme than some versions 5 The 3D 'bowl' shape was acknowledged 6 Similar to Broili (1967) 7 But recognised the limitations of this assumption 8 Rainfall influences are analysed and acknowledged 9 C3 implicit-referred to as 'strongly non-circular slide'…”

“…Despite these three-dimensional interpretations, most researchers analysed the stability of the Vaiont landslide in terms of approximately S-N cross-sections, each having a steep subplanar upper failure surface and a subhorizontal lower failure surface (e.g. Guidici and Semenza 1960;Müller 1964Müller , 1968Semenza 1965;Mencl 1966;Skempton 1966;Broili 1967;Romero and Molina 1974;Chowdhury 1978;Hendron and Patton 1985;Belloni and Stefani 1992;Hutchinson 1987;Tika and Hutchinson 1999;Kilburn and Petley 2003;; Alonso and Pinyol 2010; Pinyol and Alonso 2010; Hungr and Aaron 2013; Bolla 2012, 2015;Boon et al 2014;Del Ventisette et al 2015;Havaej et al 2015;Paronuzzi et al 2016;Wolter et al 2016;Crosta et al 2016).…”

“…Raised pore water pressures due to impoundment of the reservoir are widely regarded as the cause of the landslide (consensus point C4), whether as a first-time failure or a reactivation (Kiersch 1964;Semenza 1965;Chowdhury 1978;Hendron and Patton 1985;Hutchinson 1987;; Alonso and Pinyol 2010; Pinyol and Alonso 2010; Ward and Day 2011; Delle Rose 2012; Bolla 2012, 2015;Paronuzzi et al 2016;Havaej et al 2015;Wolter et al 2016). Unusually for the period there are some measurements of water levels and ground movements from the southern side of the Vaiont gorge prior to the landslide (Müller 1964), although these are extremely limited.…”

Mineralogical and Geotechnical Characterization of the Clay Layers within the Basal Shear Zone of the 1963 Vajont Landslide

“…This event represents a reference case history for scientists and researchers dealing with large landslides and/or reservoir-induced slope failures and is considered a milestone in the general development of rock and soil mechanics [30,31]. This famous landslide has attracted a lot of scientific interest over time, owing to its complexity and the need to understand the underlying mechanisms that caused the slope failure [21,23,26,[32][33][34][35][36][37][38][39][40]. The 1963 Vajont landslide mobilized a rock mass made up of a complex cherty and marly limestone sequence dating from Middle-Upper Jurassic to Upper Cretaceous, including the following geological Formations (Fm.…”